ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
ENEC further illustrated hardware dependence with lossless model-weight compression on Ascend NPUs.
“ENEC further illustrated this hardware dependence with lossless model-weight compression on Ascend NPUs [77].”
From Unified AI Gateway: A Framework for Joint Model Routing and KV Cache Management (2026) · cites this paper as [77]
Cites
A universal algorithm for sequential data compression
Ziv, J., & Lempel, A. (1977) · IEEE Transactions on Information Theory · doi:10.1109/tit.1977.1055714 · Cited by 5,456 · see citing works
Compression of individual sequences via variable-rate coding
Ziv, J., & Lempel, A. (1978) · IEEE Transactions on Information Theory · doi:10.1109/tit.1978.1055934 · Cited by 3,432 · see citing works
A Technique for High-Performance Data Compression
Welch (1984) · Computer · doi:10.1109/mc.1984.1659158 · Cited by 2,202 · see citing works
An Introduction to Arithmetic Coding
Langdon, G. G. (1984) · IBM Journal of Research and Development · doi:10.1147/rd.282.0135 · Cited by 526 · see citing works
Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Vaughan, A., et al. (2024) · arXiv · doi:10.48550/arxiv.2407.21783 · Cited by 505 · cited 3× · see citing works
“1-8B-Instruct [17] 16.06 8.00 ∼ 1002.00 FP16 CapybaraHermes-2.”
InfluentialSelf-citationZstandard Compression and the application/zstd Media Type
Collet, Y., & Kucherawy, M. (2018) · Request for Comments · doi:10.17487/rfc8478 · Cited by 265 · see citing works
A Study of BFLOAT16 for Deep Learning Training
Kalamkar, D., Mudigere, D., Mellempudi, N., Das, D., Banerjee, K., Avancha, S., Vooturi, D. T., Jammalamadaka, N., Huang, J., Yuen, H., et al. (2019) · arXiv · doi:10.48550/arxiv.1905.12322 · Cited by 169 · cited 1× · see citing works
“Recently, BF16 has gained popularity for large models due to its wider exponent range, which offers training stability [32], [35], [54].”
Self-citationDaVinci: A Scalable Architecture for Neural Network Computing
Liao, H., Tu, J., Xia, J., & Xi-ping, Z. (2019) · IEEE Hot Chips Symposium · doi:10.1109/hotchips.2019.8875654 · Cited by 123 · see citing works
Elf: Erasing-Based Lossless Floating-Point Compression
Li, R., Li, Z., Wu, Y., Chen, C., & Zheng, Y. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3587136.3587149 · Cited by 48 · see citing works
Knorr, F., Thoman, P., & Fahringer, T. (2021) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1145/3458817.3476224 · Cited by 29 · see citing works
The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data Only
Penedo, G., Malartic, Q., Hesslow, D., Cojocaru, R., Alobeidli, H., Cappelli, A., Pannier, B., Almazrouei, E., & Launay, J. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-3464 · Cited by 25 · see citing works
GPULZ: Optimizing LZSS Lossless Compression for Multi-byte Data on Modern GPUs
Zhang, B., Tian, J., Di, S., Yu, X., Martin Swany, A., Tao, D., & Cappello, F. (2023) · International Conference on Supercomputing · doi:10.1145/3577193.3593706 · Cited by 24 · see citing works
Self-citationFCBench: Cross-Domain Benchmarking of Lossless Compression for Floating-Point Data
Chen, X., Tian, J., Beaver, I., Freeman, C., Yan, Y., Wang, J., & Tao, D. (2024) · Proceedings of the VLDB Endowment · doi:10.14778/3648160.3648180 · Cited by 23 · see citing works
Self-citationHuffman Coding with Gap Arrays for GPU Acceleration
Yamamoto, N., Nakano, K., Ito, Y., Takafuji, D., Kasagi, A., & Tabaru, T. (2020) · Proceedings of the International Conference on Parallel Processing · doi:10.1145/3404397.3404429 · Cited by 21 · see citing works
Efficient Lossless Compression of Scientific Floating-Point Data on CPUs and GPUs
Azami, N., Fallin, A., & Burtscher, M. (2025) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3669940.3707280 · Cited by 19 · see citing works
A Multidimensional Communication Scheduling Method for Hybrid Parallel DNN Training
Li, S., Lü, K., Lai, Z., Liu, W., Ge, K., & Li, D. (2024) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2024.3406420 · Cited by 17 · see citing works
Hidayat, T., Zakaria, M. H., & Pee, A. N. C. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-14130-1 · Cited by 17 · see citing works
ZipNN: Lossless Compression for AI Models
Hershcovitch, M., Wood, A. W., Choshen, L., Girmonsky, G., Leibovitz, R., Ozeri, O., Ennmouri, I., Malka, M., Chin, P. S., Sundararaman, S., et al. (2025) · IEEE ... International Conference on Cloud Computing · doi:10.1109/cloud67622.2025.00028 · Cited by 14 · see citing works
Lightweight Huffman Coding for Efficient GPU Compression
Shah, M., Yu, X., Di, S., Becchi, M., & Cappello, F. (2023) · International Conference on Supercomputing · doi:10.1145/3577193.3593736 · Cited by 12 · see citing works
The Compression Optimality of Asymmetric Numeral Systems
Pieprzyk, J., Duda, J., Pawlowski, M. P., Camtepe, S. A., Mahboubi, A., & Morawiecki, P. (2023) · Entropy · doi:10.3390/e25040672 · Cited by 12 · see citing works
Massively Parallel ANS Decoding on GPUs
Weißenberger, A., & Schmidt, B. (2019) · Proceedings of the International Conference on Parallel Processing · doi:10.1145/3337821.3337888 · Cited by 12 · see citing works
SDP4Bit: Toward 4-bit Communication Quantization in Sharded Data Parallelism for LLM Training
Jia, J., Xie, C., Lin, H., Wang, D., Feng, H., Zhang, C., Sun, B., Lin, H., Zhang, Z., Liu, X., et al. (2024) · neural information processing systems · doi:10.52202/079017-0279 · Cited by 11 · see citing works
Self-citationLü, T., Zhong, Y., Sun, Z., Chen, X., Zhou, Y., Wu, F., Yang, Y., Huang, Y., & Yang, Y. (2023) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1145/3581784.3607044 · Cited by 10 · see citing works
Self-citationRecoil: Parallel rANS Decoding with Decoder-Adaptive Scalability
Lin, F., Arunruangsirilert, K., Sun, H., & Katto, J. (2023) · Proceedings of the International Conference on Parallel Processing · doi:10.1145/3605573.3605588 · Cited by 6 · see citing works
An Advancement in Huffman Coding With a Potential for Parallel Decoding
Iyer, K. V., Seshadri, K., & Srinivasulu, K. (2025) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70096 · Cited by 4 · see citing works
A novel method of lossless compression for 2-D astronomical spectra images
Du, B., & Ye, Z. (2009) · Experimental Astronomy · doi:10.1007/s10686-009-9169-2 · Cited by 4 · see citing works
Adaptive Encoding Strategies for Lossless Floating-Point Compression
Li, Z., Li, R., Xu, X., Wu, Y., Chen, C., Liu, T., Shang, J., & Zheng, Y. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3554999 · Cited by 3 · see citing works
Asymptotic Optimality of the Asymmetric Encoding-Decoding Scheme
Yamamoto, H., & Iwata, K. (2024) · International Symposium on Information Theory and its Applications · doi:10.23919/isita60732.2024.10858218 · Cited by 2 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Unified AI Gateway: A Framework for Joint Model Routing and KV Cache Management
Lu, J., Zhang, X., & Shao, Y. (2026) · arXiv · doi:10.48550/arxiv.2609.06940
“ENEC further illustrated this hardware dependence with lossless model-weight compression on Ascend NPUs [77].”
Self-citation
- Authors
- Yang, J., Wu, J., Liu, Z., Ma, X., Zhao, H., Gu, Y., Huang, Y., Liu, X., Huang, W., Wei, Z., Xing, J., Ma, Y., Zhang, Q., An, B., Hu, Z., Liu, S., Zhu, X., Lu, J., Tan, G., & Tao, D.
- Affiliations
- Chinese Academy of Sciences (CN)
- University of Chinese Academy of Sciences (CN)
- Huawei Technologies Co., Ltd.
- ORCID
- Venue
- 2026 ACM/IEEE 53rd Annual International Symposium on Computer Architecture (ISCA)
- Publisher
- IEEE
- Published
- 2026-06
- Pages
- 1319-1335
- ISSN
- 1063-6897 (ISSN-L)
- OpenAlex
- W7172448177
- Cited by
- 0
- References
- 26
- FWCI
- 0.00
- Type
- conference paper
- Language
- en
- Open access
- closed
- Licence
- Locations
- Proceedings - International Symposium on Computer Architecture/Proceedings (publishedVersion)
- Funders
- ICT
- CAS (E461050)
- National Key Research and Development Program of China (2025YFB3003702)
- National Natural Science Foundation of China (62032023,T2125013)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Data Compression Techniques
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Keywords
- model compression, lossless compression, inference acceleration, neural network inference
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
